1. Field of the Invention
The present invention relates to an image processing device, an image processing method, and a computer program product.
2. Description of the Related Art
Conventionally, an image processing device that processes image data obtained by scanning an image can display a preview of the image that has been scanned or a preview of the image to be processed. For example, Japanese Laid-Open Patent Application No. 2004-248057 (Patent Document 1) discloses an invention of an image forming apparatus with which the operator of the apparatus selects a finished image at a selection page for selecting the finished image, so that the copy mode corresponding to the selection is set.    Patent Document 1: Japanese Laid-Open Patent Application No. 2004-248057
However, the image forming apparatus disclosed in patent document 1 does not address the timings of generating and displaying the finished image which is a preview image. The preview image is generated by performing a zooming process to reduce the input image data, and applying other processes to the reduced image data according to need, such as a rotating process, a modifying process such as a combining process for combining two images together and to outputting them as a single image, a post process such as a punching process or a stapling process, etc. Thus, if reduced image data is generated every time a preview image is displayed, it will take a long time to display the preview image.
Furthermore, the reduction ratio used for reducing the image data varies depending on the operator's purpose of viewing a preview. For example, a rough image suffices for the purpose of confirming the vertical orientation of the image, while a relatively fine image is required for confirming the orientation of the characters included in the image. If reduced image data corresponding to an operator's request is generated every time a request is made by the operator, it takes a considerable length of time to display the preview image. Moreover, the resources of the apparatus will not be efficiently used if similar processes are repeated.
Furthermore, in the process of reducing the image data, if the zooming method were changed depending on the properties of the image, preferable results would be obtained in the reduced image data. For example, the zooming method may be changed depending on whether a region in the image is a character region or a non-character region, and whether the image is a chromatic image or an achromatic image. However, no consideration is made in this respect in patent document 1.